;/17 ~y

C &' tf7 ttJ ~ ---15"8" ~

COPY 2

MODEL INVESTIGATION OF THE

SILT EXCLUDER SYSTEM FOR THE

TRIMMU-SIDHNAI LINK

October 1962

IN DUS BASIN PROJECT

WEST PAKISTAN

Prepared for

Tipton ond Kolmbach, Inc. Consulting Engineers

Denver, Colorodo

Colorado State University Research Foundation Civil Engineering Section

Hydraulics Laboratory

CER62SSK58

MODEL I NVESTIGATION

OF THE

S ILT EXCLUDER SYSTEM

FOR THE

TRIMMU- S IDHNAI LINK

INDUS BASIN PROJECT

WEST PAKISTAN

Prepared for Tipton and Kalmbach, Inc.

Consulting Engineers De nve r, Colorado

by

S. S. Karaki

and

R. M. Haynie

Colorado State University Research Foundation Civil Engineering Section

Hydraulics Laboratory

U18401 0593851

F IGURES · · · · · · · · ·

DEFINITIONS OF TERMS

REPORT SUMMARY

INTRODUCTION · ·

THE MODEL· · · · · · · ·

MODEL TESTS AND RESULTS

Existing Conditions· · · · Schem e A (Contract Documents) · Schem e A Modification I .. ... Scheme A Modification II and III. Scheme A Modification IV . ...

CONTENTS

Extension of Excluder Tunnels to Outer Divide Wall No Exclusion Tunnel Extension . . Extension of Tunnels 1 and 2 Only . Comparative Summary . General Observations

RECOMMENDATION . . . . .

ALTERNATIVE RECOMMENDATION

ACKNOWLEDGEMENTS

BIBLIOGRAPHY

ii

iii

2

2

6

6 7 8

11 12 14 14 15 16 16

21

21

23

23

2

3

Extent of model coverage . .

Schematic of excluder model

FIGURES

Sediment excluder model sediment size distribution curve

4 View of sediment feeders used in model and screen in front of the feeders to ensure submergence of plastic particles

5

6

7

8

9

10

11

12

1 3

14

Percent of total sediment introduced in the model entering Haveli Canal

Photograph of Scheme A contract documents during test . Tunnel tops were constructed of clear plastic . . . . . . .

Left guide wall installed in accordance with Pakistan model results. Arrows show pattern of bed-load movement ......... ... .

Comparative effects of alterations to Scheme A (Cont . Documents) .

Surface flow pattern of Scheme A (Cont. Doc.) with 4 drop curtain and existing outer wall . . .

Scheme A Modification I curved tunnel top

Scheme A Modification I during test with shortened outer divide wall . The difference in tunnel top elevation between the extension and existing sluice tunnels can be seen .

Comparative results of Scheme A Modificat ion I. of sediment in canals compared to Fig. 11

Reduction

Scheme A Modification II. Tunnel extension roof raised to elevation 4 78

Scheme A Modification III. Tunnel extension roof raised to elevation 481

15 Results of Scheme A Modification II and III. Reduction of sediment in canals compared to existing condition · · . · · · · · · · · · ·

16

17

18

19

20

21

22

23

24

25

Comparative bed-load movements with Modifications II and III

Scheme A Modification IV . ....... . .. . . .

Sluice tunnels extended to include left pocket area

Results of tunnel extensions laterally to include sluices 5 to 8

Comparative results of changes in existing structure with no tunnel extension, with Scheme A contr. doc . and with Modification III

Intermediate wall removed. No sluice tunnel extension

Extension of tunnels 1 and 2 only. Tunnel No. 2 is set back 40 ft from the end of Tunnel 1. Tunnel 1 is even with the end of the T-S head regulator

Pattern of flow with existing outer divide wall and no sluices operating

Comparison of extension of Tunnels 1 and 2 only with Scheme A Modification III

Recommended silt excluder

TABLE

Summary Table of Results

ii

3

4

3

6

7

8

8

9

10

10

11

11

11

12

12

13

14

15

15

15

16

16

16

21

22

17

DEFINITIONS OF TERMS

Alluvial channel studies involve use of many terms whic h are ambiguous to the r eader and appear to have multiple meanings. The following is a list of definitions of t e rms p e rtaining to this study . All of the terms do not appear in this report but are included for purpose of clarification. The writer is aware that some of these definitions are not uni ver-sally accepted . It is anticipated, however , that there will be l ittle diffic ulty in interpret ing the con -tents of t his r eport if the definit ions give n below a r e carefully studied .

Sediment - Fragmental material t hat origi -nates from weathering of r 'ocks and is subject to transport by water.

Suspended sediment - The sediment that at any given time is moving in suspension in the water-sediment mixture above a s pecified height above the channel bed and is maintained in s uspens ion by the upward components of turbulent c urrents or by col -l oidal suspension. In this r eport the height above the channel bed is specified as O. 1 foot in the model.

Material - Connotes division of sediment by size or source (not origin) and is us ed with another t erm to designate size division; for example , coarse material or bed material.

Bed m aterial - Denotes division of sediment by sizes. In this report it includes all sediment sizes coarser than 0. 074 mm. By general definit ion it includes all sediment c oarser than the l argest stan -dard separation size at which no more than 10 percent of the bed material is finer . The standard separation size used in this report is the U. S. Standard sieve No . ZOO which has an opening of 0 . 074 mm.

Load - Connotes sediment in transport. The term should not be used interchangeably nor con -fused with concentration or discharge. The word is used to separate sediment according to mechanics of transport .

Wash load - Denotes sediment sizes transpor-t ed in suspension, and d ivision of suspended sedi -ment by sizes; in this report it includes all sediment sizes smaller than 0. 074 mm. By general definition it includes all sediment finer than the largest stan-dard separation size at which no more than 10 per -cent of the bed material is finer. The standard sepa -ration size used in this report is the U. S. Standard sieve No. ZOO whic h has an opening of 0. 074 mm .

Suspended bed material - Bed material sizes s uspended in the flow.

Bed load - Sediment that moves along essen-

iii

tially in continuous contact with the channel bed. ln this report , sediment within O. 1 foot of the bed in the model is construed as b eing essentially in continuous contact with the channel bed .

Total load - All sediment transported by the flow.

Discharge - The volume , or weight of the water, water - sediment mixture , or sediment which passes through a section of flow in a unit of time. The section may include the total section, a unit width and/ or unit depth .

Suspended sediment d ischarge - We ight of a ll the suspended sediment which passes through a sec-tion of flow in a unit of time . In this report section denotes the total cros s - section of the waterwa y.

Bed - load discharge - Weight of bed load which passes t hrough a section of flow in a unit of time . In this report sect ion de notes the total c r oss -section of the waterway .

Bed - material discharge - Weight of bed material which passes through a section of flow in a unit of time. In this report , section denotes the total c r oss - section of the waterway.

Total-sediment discharge - Weight of all sedi -ment which passes through the total cross - section of the waterway in a unit of time . It is the sum of sus -pended sediment disc harge and bed - load discharge.

Sediment concentration - The ratio of weight of sediment to the we ight of water - sediment mixture in parts per million . A part per million is a unit weight of sediment in a million unit weights of water -sediment mixture .

Bed -material concentration - Concentration of bed material without regard to mode of transport.

Suspended-bed-material concentration - Con -centration of bed material sizes in the suspended sediment .

Total sediment concentration - Concentration of sediment without regard to sizes or modes of transport.

Lower flow regime - A category for flows hav-ing bed forms of ripples , ripples on dunes , or dunes .

Upper flow regime - A category for fl ows having bed forms of plane bed with sediment move -ment , standing waves , or antidunes .

REPORT SUMMARY

A model study of modifications to the existing silt excluder of the Haveli Link Canal headworks to meet the needs of silt control at the Trimmu - Sidhnai headworks was performed at the Hydraulics Labora-tory of Colorado State University . The study was conducted for the consulting engineering firm of Tip-ton and Kalmbach, Inc . of Denver , Colorado.

The head regulator of the Haveli Canal is located on the Chenab River at Emerson Barrage immediately downstream from the confluence with the Jhelum River . The head regulator of the pro-posed Trimmu-Sidhnai Link Canal will be adjacent t o and ups tream of existing works . Preliminary model studies conducted in Pakistan indicated that with the existing sediment excluder , when the Trimmu -Sidhnai Link was installed, the sediment load enter -ing the Haveli Link could be expected to inc r ease s ignificantly . This observation was verified in the subject model study. Existing conditions in the Haveli Link are such that no appreciable increase in the sediment load is permissible and sediment load in T rimmu-Sidhnai (T-S) Link is to be minimized. Therefore , this study was undertaken in an attempt to arrive at some scheme of modifying the existing excluder system which would be satisfactory to both c anal systems .

The study was conducted in a distorted sec-tional model including approximately one-fifth of the total barrage . The horizontal scale of the model was 1: 120, model to prototype, and the vertical scale was 1 :25. The various modifications of the sediment excluder in the model were compared with the same river discharge and pond level.

Based upon the laboratory data and observa-tions of sediment movement in the vicinity of the silt excluder , assuming:

1. that the centerline of the Trimmu - Sidhnai head regulator structure, normal to the canal centerline , will be along the same line as the centerline of the Haveli regulator, and

2. that the centerlines of the two canals will be 330 ft apart ,

it is recommended that:

1. the existing intermediate divide wall be re -moved;

2 . the outer existing wall remain intact;

3 . the exclusion (sluice) tunnels should be exten-ded upstream in the manner shown in Fig . 25 . The top of the tunnel roof slab of the extension should be maintained at elevation 481 with a horizontally curved shape with a radius of 415. 79 ft as shown in the figure;

4. subdivisions of the four main tunnels in the extended section resulting in eight under-sluices will not hydraulically alter the per -formance of the silt excluder if structurally this should be desirable . If possible the main tunnels should be maintained at four to permit positive sluicing of the tunnels;

5 . the existing island immediately upstream of the silt excluder structure be removed to improve the approaching flow pattern so that much of the sediment (bed load) can be trans-ported away from the silt excluder structure and the head regulators;

6 . the left guide wall in the river approach to the Trimmu-Sidhnai head regulator, recommen-ded by the Pakistan model study, is satisfac-tory provided a wing wall at the entrance to the head regulator is constructed above eleva-tion 485. T he left guide wall shown in Fig . 7 is basically a curved wall, below the water surface extending only to elevation 485 . The wing wall at the entrance is shown in Fig . 25 .

As an alternative to the recommended silt excluder, it is suggested that simpler construction could be effected by extending under-sluices 1 and 2 only as shown in Fig . 22 . The extension of only two sluice tunnels does not create the same advantageous approach flow pattern as the recommended structure with regard to transporting bed load toward the bar -rage. Sediment will tend to deposit upstream of sluices 3 and 4 and probably will require more fre -quent flushing than the recommended structure to minimize sediment entry into the links . The chief difference in location of sediment deposit between the recommended structure and the alter-native would mean less sediment entry into the canal.

INTRODUCTION

The Trimmu - Sidhnai {abbreviat ed her eafter as T-S) link canal will be constructed with the head regulator adjacent to the existing Haveli Canal. Be -cause of desirable operational procedures of the canals, although t he two canals are adjacent to each other, a separate he ad regulator will be construct ed for the new T-S link . Increase in offtake discharge from the river requir es a lteration of the head works , specifically the silt excluder , if sediment problems are to be minimized in both canals . A greater off -take discharge generates greater local approach velocities in the river whic h in turn will transport greater quantities of sediment into the exist ing Haveli Canal unless a structure is constructed to reduce sediment entry into the canals.

Some s tudies conduct ed in Pakis tan in a small scale model of the entire Emerson Barrage and the

approach r iver section on a l :250 horizontal scale (model to prototype) distorted vertical scale model indicated that unless s ome specific alteration was made to the existing excluder structure the Haveli sediment load would increas e very appreciably by addition of the T-S canal.

A sect ional model of only the head regulators and approximately one - fifth of the total Emerson Bar -rage was constructed at the Hydraulics Laboratory of Colorado State University to s tudy in greater detail the silt exclusion problem for the combined Have li -T-S headworks to arrive at a favorable solution for minimizing sedim ent (bed - material load) entry into the two canals . In general, the objective of the study was to maintain or possibly minimize the sediment load in the Haveli Canal compared t o existing conditions and t o develop an effective excluder for the c ombined head works .

T HE MODEL

Distortion of a river model i s often a neces -sity . By geometric distortion it becomes poss ible to effectively r eproduce turbulent flow conditions which would not' easily be possible to attain in a comparably sized undistorted model. Distortion of the model permits greater flow velocities necessary for m ove -ment of sediment. When a model is geometrically distorted , unavoidably other distortions are intro-duced. In order to r eproduce prototype conditions as nearly as possible it is often expedient to make the other necessary adjustments through trial and error procedures.

A di s torted mode l was deem ed necessary for this s tudy . The horizontal scale of 1: 120 (model to prototype) was det ermined on the basis of available space and necessary phys i cal coverage of the river . The vertical scale was then selected so that depth , and subsequ ently velocity of flow in the model would be sufficient to creat e favorable flow velocities for movement of sediment.

A sectional model was construct ed , primarily because of limited space . Ho wever, since the prob -lem was localized at the canal head regulators , it was deemed prudent t o construct only that portion of the river , and canal head works inc luding Emerson Bar-rage necessary for the s tudy , particularly since a compl et e model of the river and divers ion works had been construct ed and were under t ests in Pakis tan. The tests in P akis tan included other phases of the total problem. The limits of the model are shown

2

as a portion of the general plan of the site in Fig . 1.

In order to control flow direction in the sec -tional mod el a syst em of gates and baffles was devised so that the discharge through each sect ion of the model could be regulated independently . There was no s pecific effort to reproduce flow patt erns of the prototype exactly because (a) no prototype data were available with regard to flow velocities or direc -tion , (b) discharge distributions around the island were not known , and (c ) barrage gate regulation pro-cedures were not s pecified. An attempt was made however to establis h flow direction in accordance with pictoral views of the Pakistan model for similar discharges us ed in this model.

The various excluder schemes were studied by comparison with the existing structure and with other excluder schemes . Results are comparable only at similar dischar ges and although flow dist ribu -tion into t he model at a particular discharge was maintained the same for the various schemes , flow patterns in the model varied because of the effects of the vari ous s tructural alterations in the vicinity of the excluders . The alterations includ ed removing of exi sting divide walls , removal of the island , intro-duction of a ne w left gu ide wall and introduction of the Trimmu- Sidhnai head regulator . A schematic plan of t he model is shown in Fig . 2 .

T he model was constructed with pumped re -circulat ed clear wat er . The bed of the model was

Lu

CHENAB RIVER

l1sLAN_D __ _

I

LIMITS OF I I MODEL "'-...J ISLAND

L __ _

FIG. I EXTENT OF MODEL COVERAGE

7 I

,/ (

HAVEL! CANAL

,:.

STILLING BASIN

A

ISLAND BAFFLE')

Q

I I

I , I

I f T-S

, I

I

, I ,

HAVEL! CANAL

CANAL 1

( I

I

HAVEL! CANAL r:r-------- --------------------lNTA_!§ EL . 48 1 1111~~------------.

EL. 468

SECTION A -A

FIG. 2 SCHEMATIC OF EXCLUDER MODEL

A

01

U>

1.0

0.8

0.6

0.4

rn 0.3 0 ~ fTl z -i 0.2 U> N fTl

-z 3 3

0.1

0.01 0.01

~ ~

l,...----'" ~ ~

~ ~ ~

~ --~

~ /

/ /

0.1 1.0 10 20 30 40 50 60 70 80 90 95 99 99.9

PERCENT FINER

FIG. 3 SEDIMENT EXCLUDER MODEL SEDIMENT SIZE DISTRIBUTION CURVE

formed of fine sand having a median diameter of O. 23 mm. A size distribution c urve for the sand is shown in Fig. 3 . Although the quantity of sand movement was purposely designed to be small , the sand bed pro-vided a moveable boundary and local points of scour and deposition were better identified . Plas tic pellets, cylindrical in shape , 1/8 11 in diameter and 1/8 11 long were used to represent the bed material l oad in the river . The specifi c gravity of the pellets was about 1. 05, varying between 1. 04 and 1. 06 . Because of the light weight of the plastic particles, some floated on the water s urfac e due to surface tension, other par-ticles remained suspended in the flow and portions were trans ported as bed load. The plastic pellets were introduced into the model by regulated mechani -cal feeders shown in Fig . 4 and reclaimed down-s tream of the model from a screened trapping device. To prevent excessive quantities of the plastic parti -cles from floating , the pellets were soaked in an off-line container for a period between 48 and 72 hours before use. Reclaimed pellets were re-admitted t o the soaking vats for re-use . Even with thi s precaution preliminary t est s resulted in excessive floating par-ticles because individual pellets dried out while in the hoppers of the mechanical fe eders . The problem was satisfactorily solved by placing a screen in front of the line of feeders as seen in Fig . 4, and applying a mist of spray of water upstream of the screen. The screen necessitated the pellets to flow below the water surface and the spray wetted the pellets up-s tr eam of the screen to ensur e submergence .

Flow into the model was measured by an orifice meter, and flow out of the model , divided into five separate channels shown in Fig. 2 was measured over V-notch weirs . This enabled separ -ate control and measurement of flow through the T-S head regulator, the Haveli head regulator, the four under sluice gates , the four gates adjacent to the under s luice tunnel and the flow through the barrage . The gates in each of these segments were raised equally as possible to prevent arbitr ary flow distri -bution in the river approach not directly comparable from one t est to another. The quantity of sediment passing through each of the fiv e flow channels were collected individually and weighed . The amount of sediment flowing into the canals was expressed as a percentage of the total sed iment c olle cted during the t est period . The effectivenes s of the individual excluder schemes were det ermined by comparison with other sc hemes.

Fig . 4. View of sediment feeders u sed in model and screen in front of the feeders to ensure submergence of plastic particles .

Three river discharges were represented in the mod e l at 75 , 000, 150,000 and about 250,000 cfs . T o det e rmine model discharges, it was necessary to assume that total river flows , less the off-take discharges , flowed over the barrage with the left and right under sluices and pockets normally closed . One - fifth of this flow plus the flow in the Haveli and when applicable the Trimmu-Sidhnai Canal , was added to make up the model flow . P ond levels were varied initially between elevation 487 and 493, how-ever , most comparative results for the various excluder schemes were made with pond level at elevation 490. Canal discharges were also initially varied, however when significant differences were not evident between different canal discharges , the discharges were set at 7000 cfs fo· Haveli ? ild 12,000 cfs for the T-S link.

MODEL TEST S AND RESULTS

In this study a number of different sediment excluder schemes were investigated. Some of the minor c hanges were studied and discarded after observation . All of the major changes , however , were studied in detail.

Exi sting Conditions

To provide a basis for c omparing the effec-tiveness of the various excluder schemes a sequence

6

of tests were made with existing river and headworks conditions . River discharges of 75 ,000, 150,000 and 250,000 cfs were t est ed with pond levels at elevations 487, 490, and 493. For each combination of these variables, the discharge in the Haveli Canal was varied at 7000, 5000 and 3000 cfs . The amount of sediment flowing into the Haveli canal based upon the total amount of sediment flowing through the model in the same test per iod is presented in the form of percent in the bar c harts of Fig. 5. A photograph

EXISTING CONDITIONS

RIVER DISCHARGE IN C.F . S 75 ,000 150,000 250,000 20-+----------------------_;_ ___ _._20

f-z w ~

r<10r<lOl'- l'-r00 l'-enenenencocoenen co <:t <:t <:t <:t <:t <:t <:t <:t <:t

r<J01'-r<J01'-r00r,.. en en co en en co en en co <:t <:t <:t <:t <:t <:t <:t <:t <:t

r<l 01'- r<l O I'- r<l O l'-en en co en en co en en co <:t <:t <:t <:t <:t <:t <:t <:t <:t

_j a: 15

_J

0..

10

5

0 0

FIG. 5 PERCENT OF TOTAL SEDIMENT INTRODUCED IN THE MODEL ENTERING HAVELI CANAL

of a t ypical t est is shown in Fig. 4 and a drawing of the existing conditions is shown in Fig. 1. The test data of all comparative schemes are t abulated in a summary table at the end of this section.

Scheme A (Contract Documents)

Scheme A refers to the plan for silt exclus ion shown in t he contract docume nts prepared by Tipton and Kalmbach , Inc . as drawing No. S2-39 . It is redrawn in this report in Fig. 2 s hown in dashed lines. A photograph of the model for this scheme i s shown during a t est in Fig . 6 . The center line of the head regulator for the T-S link is along the same centerline for the existing Haveli head regulator wit h the distance between centerlines of canals set at 330 ft. The only difference between the model scheme and that s hown in the contract documents is that the top of the intermediate divide wall was lowered to elevation 481 instead of e l evation 483 as shown in the docume nt s . T he existing sluice tunnel s are extend ed

7

to meet the head regulator of the T-S canal with additional internal tunnel divide walls to create 8 tun-nels where formerly there were four.

From observation of the results of tests with the existing condition, it was decided to make com -parison tests of the various schemes wit h river dis-charge of 1 50 , 000 cfs and pond level of elevation 490. Have li discharge was set for 700 0 cfs and the Trimmu -Sidhnai discharge at 12,000 cfs.

The te s t s were run first with the island in place as it presently exists but with the flow obst r u c -ted to the l eft of the is land. Additional tests were then performed with the island removed and with a left guide wall as shown in Fig. 7. The firs t series involved tests with the outer divide wall undisturbed and then with the outer divide wall r educed in length to 155 ft from the weir centerline . T he second series involved an additional t est with the under sluice

Fig . 6 . Photograph of Scheme A c ontract documents during t est. Tunnel tops were constructed of clear plastic.

discharging 5000 cfs. T he var i ous amounts of sediment entering both c anals are shown in bar graph form in Fig . 8.

The island very definitely formed an obstruc-tion to the canal approach . It would appear not only desirable but nec essary to remove at least part of the island to develop a more favorable approach flow pat-tern .

T he left guide wall to the head regulators as developed in the Pakistan model and tested in the second series indicated a very definite change in the flow line approach to the head regulators. By creat-ing a greater flow approach area , the velocities in the vicinity of the outer divide wall were reduced , less sediment was suspended , and thus l ess material was transport ed into the canals . It was evident t hat a proper approach to the head regulators is a factor to be considered integrally with the silt exclud er struc -ture itself . With the island in place , the percent of total model sediment entering Haveli amounted to aboot 4 percent , while wit h an improved flow approach , the sediment entry was reduced to about

8

' \ \ ' ' I t I '( I

I I I I ; I

Fig . 7. Left guide wall installed in accordance with Pakistan model results . Arrows show pat tern of bed - load movement .

3. 5 percent . Sediment entry into the T - S canal how -ever was not materially altered . A decrease in le ngth of the outer divide wall served to increase the sediment entry into Haveli by about 1 percent and decreased sediment entry into the T-S link by about the same amount . Operat ion of the sluice tunnels increased sediment entry into both canals by about 1/2 percent , resulting largely from the greater approach flow velocit ies in the river which s uspended a greater amount of sediment in the flow . In all tests the sediment entry into Haveli increased slight -ly in comparison to the tests with exi s ting c onditions .

A curtain wall, 4 ft in height installed across the openings of the tunnels from elevation 48 1 to elevation 477 proved tor-educe the amount of sedi-ment enter ing both canals. The effect of the curtain wall was to provide a flow deflector which created a pronounced c urvature in the approach flow and with it development of a strong secondary current which carried the bed load away from the head regulator toward the barrage at the right of the outer divide wall. A photograph which attempts to show the sur -face flow pattern is shown in Fig . 9 .

Scheme A Modification I

15

>-z w ~ 0 w _J (/) <t. _jz 10 W<t, gu ~(/) _J 0:: 4.W >->-oz f-W

u.I Ou 5 >-I ~3: u 0:: w a.

0

SCHEME A ( Contract Documents)

RIVER DISCHARGE - 150,000 C.F.S.

:J :J :J w (/)

> I w (/) w (/) <t. :J :J > > I >- ::J :J -_J I I -w (/) w (/) <t. <t. w (/) w (/) w

I >-

.

--

> I > I I >- > I > I > <t. <t. <t. <t. <t. I >- I >- I >- I >- I

I A 8 A 8 C D E EXIST.

INI SLt,NPn: iSU ,ND REM~VED GUIDE W LL INSTALLED

COND.

EXPLANATION OF CODE LETTERS

A. Existing outer divide wall 8 . Outer div ide wal I reduced to I 55 ft . C. Outer wall reduced to 155 ft.

Sluices I - 4 discharging 5000 c.f.s. D. 4 feet drop curtain across tunnel openings

with existing outer divide wall E. 4 feet drop curtain with 155 ft. outer wall

15

L- 10

.... 5

0

FIG. 8 COMPARATIVE EFFECTS OF ALTERATIONS TO SCHEME A ( Cont. Documents)

This scheme differs from the contract documents only in the shape and elevation of the tun-nel extension roof slab a s shown in Fig. 10. While t he existing roof slab of the sluice tunne ls is at elevation 481 this modification lowered the extension to elevation 4 7 5. A photograph of the model for this scheme is shown in Fig . 11,

T he modification r esulted in reduced sediment entry into both canals as indicated in the bar graph Fig . 12. T he improvement was largely due to re-duced velocities above the tunnel roof in the immedi -ate vicinity of the T-S head regulator , causing less suspension of bed material and less carry- over into both canals . The extr emities of the sluice tunnels with respect to the outer divide wall also was signifi -cant in that it permited the approach flow to be

9

Fig . 9 . Surface flow pattern of Scheme A (Cont . Doc .) with 4' drop curtain and existing outer wall .

diverted causing greater flow curvature and stronger secondary current. Wit h a stronger secondary cur -rent more bed load was t ransported over the barrage away from the head regulators . Reduction in t he length of t he outer divide wall resulted in slightly in-creased sediment load into Havel i and reduced sedi -ment load into the T - S canal.

(t. T-S

With the e levation of the tunnel extension at 475 , t here was a g r eat er t endency for deposition of sediment on the tunnel roof s lab, which in operat ion may require some method of det er mining when sluicing would be necessary. If t he sediment is allowed to deposit in quantity above the extended sluice tunnels , the sediment load into the canals would increase . The difference in elevations bet ween the exist ing tunnel roof s lab and the tunnel extens i on namely 6 ft from 48 1 to 4 75 , would enable sluicing of the deposit on the tunnel extens i on . Sluicing , that deposit of sediment however , would not be completely effect ive because the area of the roof slab and dis -tance from the open ing at the beginning of the tunnel extension is large .

Some modifications invo lving upstream bed vanes , and surface vanes were attempted but from visual observations and comparisons it was con-cluded that vanes of either type would not be materially effective in directing t he bed material away from the head regulators . In fact in some cases sediment l oad into the canals increased be-cause of the increase in turbulence over , around and under the vanes , which creat ed greater suspension of sediment in the flow .

One of the alterations t o t he basic scheme att empted was to place a s ill on the extension roof slab t o a height of elevation 48 1 as shown in Fig . 10, with the hope that this would tend t o increase the secondary circulation of flow in the vicinity of the silt excluder, hopefully to direct more of the bed l oad over the barrage . T he result however was to in-crease the sediment into both the canals because t he sill created turbul ence and mat eria l whic h norm a lly wou ld have deposit ed on the roof slab was suspended in the flow and carr ied into the canals . T he effect is shown in Fig . 12 C as compared to A .

<t HAVEL 330' --

- -~:_EL.475 , - r-- ---POSITION OF--- - , ,' -EL.481 ·,

TUNNEL NO. I 2 3 4

FIG. 10

SILL -- ... ,,: 7 - ---;+ -- ~- - - - -- -

/ ,,:r--' ...... ----...j '- .,,.

SLUICE TUNNELS

OUTER DIVIDE/ WALL

SLtJ ICE NO. 5 6 7 8

SCHEME A MODIFICATION I CURVED TUNNEL TOP

10

Fig. 11 . Scheme A Modification I during test with shortened outer divide wall. The difference in tunnel top el evation be -t ween the extension and existing sluice tunnels can be seen .

Operation of the s luice tunnels did not effec-tively reduce sediment entry into the canals. As in the tests of Scheme A , contract documents, the

increased velocities of approach in the river produc ed increased sediment entry into the Haveli Canal.

Scheme A Modifications II and III

T he tunnel extension roof of Modification I of Scheme A was raised successively in Modifications II and III to elevation 4 78 and 481 r espectively. Photo-graphs of the model arrangements are shown in Figs . 13

Fig . 13 . Scheme A Modification II. Tunnel extension roof raised to e levation 4 78.

SCHEME A MODIFICATION I

1-z w ~

RIVER Q :: 150,000 c.f.s. POND LEVEL = 490

0

EXPLANATION

A. Exist ing outer divide wall B. Outer wall length 155 ft. C. Curved sill on tunnel extension

roof to elev. 481 existing wall D. Condition C with sluices I - 4

discharging 5000 c.t s. E. Condition A wit h sluices 1-4

discharg ing 5000 c.f. s.

FIG. 12 COMPARATIVE RESULTS OF SCHEME A MODIFICATION I. REDUCTION OF SEDIMENT IN CANALS COMPARED TO FIG. 11

11

and 14. Modification II r esulted in reduced sediment entry into both canals when compared with Modifica-tion I. The sediment entry into Haveli was less than that recorded for the existing condition . The pe rcent of sediment into the T-S canal was approximately twice that in Haveli , or about 2 to 2 , 5 percent of the total sediment introduced into the model. T his c om -pares with about 2 percent entering Haveli with the existing conditions .

The basic diffe r ence between Modifications II and III was that Modification III resulted in greater flow deflection at the silt excluder , creating in turn greater secondary circulation , T his secondary c ir -culation directed more of the bed mat erial load in the river away from the head regulators toward the bar -rage . The results of the modifications are depict ed in the bar graphs of Fig , 15 . The left guide wall was installed as recommended from the Pakistan model s tudy . Reducing the length of the outer divide wall to 155 feet increased the sediment load into both canals . Opening either set of sluice gat es , 1 to 4 or 5 to 8 served only to increase the sediment into the canals . Under these circumstances it may be advisable to reduce the discharge rates in both canals while sluicing deposits of sediment from the vicinity of the exclusion structure .

Figure 16 shows schematically the difference in the e ffect of the height of the tunnel extension on the deflection of bed load toward the barrage . The higher e l evation of Modification III defle cted the bed load toward the barrage from a point further upstream than Modification II. T he solid arrows indicate the trend in bed - load movement with Modification III.

Scheme A Modification IV

Fig. 14 , Scheme A Modification III. Tunnel extension roof raised to elevation 48 1,

A seri es of test s were conduc t ed with the centerlin e of the T - S head regulator set back 50 ft from the head regulator of Have li with the basic

I-z w :::E C) w_J en <1. _jz W<t. gu ;:Een a:: _jW ~I-Oz 1-w LL :C og

:c ~~ w (.) a:: w Cl..

10 9

8

7

6 5 4

3 2

0

SCHEME A MODIFICATIONS

RIVER Q = 150,000 CF.S. POND LEVEL 490 ft.

MODIFICATION II. MODIFICATION m en -

-3: en ~ I -- I--

-ci en en en en en en en > I I I :c I :c I :c I :c :c :c :c :c I :c

_<( I- I- I- I- I- I- I-:c ---- I I I -=• I

A B C D A B C D E EXIST ISLAND REMOVED GUIDE WALL INSTALLED COND

EXPLANAT ION :

A. B.

C. D.

E.

Basic modificat ion with exist ing outer divide wall

Outer div ide wal I reduced to 155 ft .

Ex ist ing outer wall with sluices I - 4 discharging 5000 ft.

Existing outer wal I with sluices 5 - 8 discharging 5000 ft. sluices I -4 closed Lowered outer wal I to EI. 485

-10

9 8 7 6 5 4

3 2

-------- I

0

FIG. 15 RESULTS OF SCHEME A MODIFICATION II. AND ill REDUCTION OF SEDIMENT IN CANALS COMPARED

TO EXISTING CONDITON

12

..... w

T-S HAVEL I WING WALL ABOVE

EL. 485 --- -,

- - - - -·--=---_-=-_ =--~ I I I I I .,,, --, ,,,, I r,' / R= 1000 - ::.--- - ~ \

,,,,,. ' ,,, .,. --~ - - ' .... _,,,, _...,,,, l ~ ~ \ - - _,, ---·-- ........ - --- ', ----... ~ "' ----~ ------~ EXISTING / - WALL:' -.......

........ , '""

- - - - - BED LOAD MOVEMENT MODIFICATION TI BED LOAD MOVEMENT MODIFICATION ill

FIG. 16 COMPARATIVE BED LOAD MOVEMENTS WITH MODIFICATIONS II. AND ID. NOTE THAT BED LOAD TURNS FARTHER UPSTREAM WITH MODIFICATION Ill BECAUSE OF GREATER DEFLECTING EFFECT

OF THE HIGHER TUNNEL TOP ELEVATION

Modification I of Scheme A. Although the sediment percentages into the canals are not directly com-parable to Modification I because the is l and was not removed , visual observat ions indicated no signifi-cant improvement with the head regulator set back . The slightly greater approach flow area created by the set back of 50 feet would create slightly less river approach velocities , but the effect on deposi-tion of suspended sediment would not be s ignificantly different fr om say Scheme A Modification I , II or III.

T he results of this series of tests are compar -able to the t ests of Scheme A , c ontract documents , (Fig . 8) with the island in place. In this comparison there was litt le or no significant difference in the per -centage of sediment entering the Haveli while sedi-ment in the T-S increased markedly . T he sediment into Haveli was approximately twice that of the existing c onditions .

In an attempt to increase t he flow area im-mediately upstream of t he T-S head regulator , thus reduce velocities and increase sediment deposition there , t he tunnel extension was removed . As the result in Fig . 1 7 shows, they did not decrease sedi-ment entry into the canals significantly . T he opening in the sluice tunnel created by the difference in eleva-tions of the existing tunnels and the extensions were closed off. This did not decrease sediment in the canals . Observations of sediment movem ent in t he model during these tests t ended to confirm that t he greatest silt exclusion coul d be effected by creating a favorable curved fl ow pat tern in t he river approach , with as simple an exclusion structure as possible to

SCHEME A RIVER Q = 150,000 c.f.s.

POND LEVEL = 490 f t.

13 MODIFICATION ISl

I- 12 z 11 w en :!!: I en en 0 10 en I-w _; I I I I I I en <1. 9 I- I-z I-_J <1.

8 W U Oen Oa: 7 :!!: w _J ~ 6 ~w 5 ~ I

(_) 4 ~:i: 3 ~ I-z 2 ~ a: I ~ 0

A A B C EXIST COND

prevent flow disturbance and unnecessary suspension of bed material which could be conveyed into the canals.

Extension of Excluder Tunnels to Outer Divide Wall

T he excluder tunnels were extended laterally across sluice gates 1 through 8 to t he outer divide wall . The change is shown pictorially in Fig . 18, The top of the tunnel slab was set at elevation 48 1. This scheme resulted in f avorable movement of bed load toward the barrage in the river approach channel but did not compare favorably with canal sediment load of Scheme A Modification III. T he percent results of the tests are given in Fig . 19. Because of the lack of improvement over , say , Scheme A Modifi-cation III the larger structure in comparison is un-justified .

No Exclusion T unnel Extension

T ests were conducted with a structural arrangement of the excluder involving removal of the intermediate divide wall and no sluice tunnel exten-sion beyond the exist ing length. T he results are shown in Fig . 20 . Without the tunnel extension the curved flow pattern and favorable secondary circula-tion near the head regulators was not pronounced and the pattern of bed - load movement indicated the change . Sediment load in both canals increased. T he photo-graph of Fig. 21 , t aken during a t est run with t he existing length of outer divide wall shows , by arrows , that t he bed load was transported past the T-S head regulat or and almost to the exist ing sluice tunnels before turning . This path or pattern provides m ore

EXPLANATION : A. Exist ing outer divide wall

8. Tumel extension top removed

C. Tunnel extension top on intermed iate opening closed

FIG. 17 SCHEME A MODIFICATION N. T-S CANAL SET BACK 50 ft.

14

opportunity for transport of bed material into the canals than the pattern creat ed by Scheme A Modification III.

Extension of Tunnels l and 2 Only

In the interest of providing t he least s truc -ture necessary for the greatest s ilt exclusion pos -sible , tunnels 1 and 2 only were extended in a manner s hown in Fig . 22, Tunnel 1 was ext ended to the upstream , or left side of the T-S head regulat or and T unnel 2 was 40 ft shorter , The tops of both tunnels were at e levation 481, the same as the exist-ing level.

Fig , 18 , Sluice tunnels extended to include left pocket area .

The movement of bed load away from the head regulators was not so pronounced as in Scheme A or its modifications and as a re sult sediment entry into Haveli was greater than the previous schemes , At the same time , sediment entry into the T-S Canal was less , as less turbulence was created near the entrance to this head regulator . The pattern of bed -load movement is s hown in Figs . 22 and 23. The

I-z w ~ 0 8 W_1 U)<f

7 _Jz w<f 6 ou OU) ::Ea:: 5 _Jw 4 <fl-t-z

3 ow I-I 2 LL(.) O:i: t-3: z 0 w (_) a:: w 0...

RIVER Q = 150,000 c.f.s . _J <f

POND LEVEL = 490 ft . z _l<f

WU 6 6 EXPLANATION: OU) O a:: U) U) U) U) ::Ew 5 I ~ I I I t'..-I I I 5 A. Existing outer divide wall

I- I- I-_lz 4 4 i=!W B. Sluices I - 4 discharging 0I 3 3 5000 cJs. t-S2 u..I o3: 2 2 C Sluices 5 - 8 discharging 1-I- I I 5000 c.f.s. zZ I - 4 closed wW 0 0 u::E a::- A B C A EXIST WO 0... t7l COND

FIG. 19 RESULTS OF TUNNEL EXTENSIONS LATERALLY TO INCLUDE SLUICES 5 TO 8

RIVER Q = 150,000 c.f.s. POND LEVEL = 490 ft.

SCHEME A NO TUNNEL EXTENSION MOD. cgNT. m oc. - en

EXPLANATION 8 A. Existing outer divide wall

- =- I -en en en f-I I I

- ...J f- f- f- en -wen I I I I I I I

- >, f- -~f-- -

7 B. Outer wall 155 f.t. 6 c. Ex ist ing outer wall with 5 sluices 1-4 discharge = 4 5000 c.fs.

- - 3 D. Exist ing outer wall wi th

- I -- I -2

sluices 5-8 discharge = 5000 c.f.s.

0 A B C D A A EXIST

ISLAND REMOVED COND. GUIDF WAI I INSTALL i;:-r,

FIG. 20 COMPARATIVE RESULTS OF CHANGES IN EXISTING STRUCTURE WITH NO TUNNEL EXTENSION, WITH SCHEME A CONTR. DOC. AND WITH

MODIFICATION ill

15

Fig . 21. Intermediate wall removed. No sluice tunnel extension .

Fig . 22. Extension of tunnels 1 and 2 only. Tun-ne l No . 2 is set back 40 ft from the e nd of Tunnel 1. T unnel 1 is even with the end of the T - S head regulator .

scheme was not as favorable as Scheme A Modifica-tion III but the slight increase in sediment l oad into Haveli which is offset by the reduction in cost of con-struction may be worth consideration as an alterna-tive to Scheme A Modification III . The results are graphically shown in Fig . 24 with a comparison to Scheme A Modification III .

Fig . 23. Pattern of flo w with exist ing outer divide wall and no sluices operating .

Comparative Summary

The results of all of the various schemes and modifications are tabulated for ease in comparison and reference is made directly to the various s c hemes. In addition to results previously shown in graphical forms, other test results are 'also given in the table .

General Observations

The results of this model study should not be attempted to be scaled to prototype . The model was d esigned and tested to compare relative improvement or advantages of various excluder structures and head r egulator positions . Further , the test results of dif -ferent river discharge condition s are not comparable one from another , for the sediment quantities in this sectional model are not necessarily in proper propor -tion to the prototype. T he recommended silt excluder is based as much upon observation of flow conditions and sediment movement as it is on recorded data of the kind shown in the figures and table of this r eport .

It is to be noted in comparing the results of tests with different river discharges that the percent -age of sediment entry into the canals at 75,000 cfs river discharge is great er than with river discharge of 150 , 000 cfs. This is misleading because it is known that sediment quantity ent ering the canals with l ower river d ischarge is l ess than at higher river dis -charges . Actually the quantity of sediment introduced into the model at the representative river discharge of 75 , 000 cfs is greater than its proportionate amount when compared to 150,000 cfs . Thi s was purposely done in this model to enable better sediment measure -ment in the canals.

16

SUMMARY TABLE OF RESULTS

Water Total ilodel Sediment Discharge feecl rate of in % of

Pond in 1000 cfs Sediment Total Model See Scheme level River Haveli T-S II/hr Haveli T - S Remarks Fig .

Existing 493 75 7 - 35 4 . 3 - 5 Conditions 490 7 - 1 1. 4

493 5 2 . 9 490 5 10 . 0 487 4 . 55 1 3. 0 487 4 14 . 3 493 3 1. 4 490 3 8 .6 487 2 4 . 6 493 150 7 - 295 2 . 0 - 5 490 4 . 0 487 2 . 0 493 5 1. 7 490

I 2 . 0

487 1. 5 493 ! 3 1. 4 490 i 1. 4 487 I o. 7 493 250

I 7 - 11 00 1. 8 - 5

490 1 . 3 487 6 1. 0 493 5 1. 6 490 ! 1. 1 487 4 1. 1 493 3 0. 4 490 0 . 8 487 2 1. 5

A Contract 490 150 7 12 35 4.0 6 . 5 Existing outer wall 8 Docum nts 4 . 5 5 . 5 155' long outer wall

3. 5 7. 0 Island removed Existing wall

4.5 8 .0 155' long outer wall Island removed

4 . 0 9 . 0 Same as above Sluices 1- 4 Q = 5000 cfs

2 . 5 4 . 0 Curtain wall 4' acrosi tunne l openings Existing wall

2. 7 4 . 0 Same as above 155 I outer wall

A Modification I 490 150 7 12 40 2. 5 4 . 5 Existing outer wall 12 Rounded tunnel I 3. 0 4.0 155' long outer wall Extension with I 3. 2 7 . 0 Curve d sill on top of top elev . at tunne l Ext. 475. See Fig. 10 I 3. 6 4 . 0 Sa rn e as above with

Sluices 1-4 Q = 5000 cfs

3. 0 3. 5 Sill r e move d with Sluice s 1 - 4 Q = 5000 cfs

17

SUMMARY T ABLE OF RESULT S (Continued )

Wa.t e r T otal Model Sediment D isc harge feed rate of in "/, of

:,ond in 1000 cfs Sedime nt Total '.Vlodel See Sche m e evcl River Haveli T - S #/hr Haveli T - S Re m arks Fig

A Modifica - 49 0 150 7 12 40 1. 0 2 . 5 Existing ou ter wall 15 tion II 1. 7 3. 3 155 ' lo ng outer wall

Tunnel extension 3 , 2 2 . 8 Existing oute r wall roof rai$ed to Slu ices 1- 4 e l ev . at 4 78 Q = 5 000 cfs

6 . 0 4 . 3 Existing oute r wall Sluices 5 - 8 Q = 5000 cfs ,.-

A Modifica -tion III 490 150 7 12 43 1. 3 2 .2 E x isting outer wall 15

4 . 6 4 . 3 155 ' long outer wall 2. 1 7 . 6 E xisting outer wall

Sluices 1- 4 Q = 5 000 cfs

3 . 1 8 . 4 E x i s t ing outer wall Sluices 5 - 8 Q = 5 000 cfs

5 . 0 3. 1 Lowere d outer wall to E lev . 485 .

A Modifica - 490

I 150 7 12 4 0 3. 5 1 3. E xisting outer wall 17

tion I 4. 0 6 . 0 Sluices 1-4 wit h T - S he ad Q = 5000 cfs r e gulator set 3.5 5 .0 Sluices 1-4 back 50 ft . Q = 12,000 cfs

5 . 0 5 .0 Sluices 5 - 8 I

Q = 10,000 cfs Sluices 1-4 clos ed .

49 0 150 7 12 35 3. 0 7. 0 Tunne l Top re m ove d

I Ends of form er tunne l ope n ings

close d. 3 .4 10. 0 Same as above wit h

155' oute r wall 4 . 0 10 . 0 Sa m e a s above wit h

Sluice s 1 - 4 I Q = 50 00 cfs

4. 0 7 .0 Tunnel Top r eplaced , e nd s ope n , i n ter -m ediate openings clos e d .

T - S Head Regu - 49 0 150 7 12 40 2 . 8 3. 7 E xi s ting outer wall 19 l ator in li ne 2 . 6 3 . 2 Sluices 1-4 L ateral Exte nsion Q = 5 000 cfs of Tunnels . 3 . 3 4. 3 Sluices 1-4 closed Roof Elev . 481 Sluices 5 - 8 I

Q = 5000 cfs No T unnel 4 9 0 15 0 7 12 4 0 3. 7 7. 3 E x is ting outer wall 2 0 Extensions 3 . 3 5 . 1 155' l ong outer wall

3 . 3 5 . 2 Exis t i ng outer wall Sluices 1 - -! Q = 5000 cfs

2 . 8 5 . 0 Existing outc r wall Sluices 5 - 8 Q = 50 00 cfs

18

SUMMARY TABLE OF RESULTS - Continued:

Water Total Model I Sediment Discharge feed rate of in % of

Pond in 1000 cfs Sediment Total Model See Scheme level River Haveli T-S fl/hr Haveli T-S Remarks Fig.

Extensions of 490 150 7 12 40 3,3 2. 1 Existing outer wall 23 Tunn ls I a nd 2.9 3. 0 155' long outer wall 2 only 2.2 4.4 Existing outer wall

Sluices 1-4 Q = 5000 cfs

3 . 1 2 . 8 Sluices 5 - 8 Q = 5000 cfs

A :viodifica- 490 150 7 12 40 1. 5 3.2 Existing outer wall tion II 1. 7 3.3 155' long outer wall

Repeat 2.5 3. 7 155' wall with sluice: 1-4, Q = 5000 cfs

I 490 75 7 12 15 37 14 Existing outer wall 13 18 155' long outer wall 40 6 . 5 Existing outer wall

with Sluices 1-4 Q = 5000 cfs

A Modifica - 490 150 7 12 42 1. 1 10 Existing outer wall tion III 35 2 . 9 13 155' l ong outer wall

Re peat 33 3. 3 10.2 Existing outer wall Sluices 1 -4 Q = 5000 cfs

40 2.5 9.6 Sluices 5-8 Q = 5000 cfs

490 75 7 12 9 13. 3 46 Existing outer wall 13 11. 3 28 155' long outer wall 11 27 5.2 Existing outer wall

with Sluices 1-4 Q = 5000 cfs

17 7.2 0.9 Existing outer wall Sluices 5-8 Q = 5000 cfs

490 2 30 7 12 40 1. 7 4.4 Existing outer wall 2.5 5 . 1 155' long outer wall 2.4 9 .6 Existing outer wall

with Sluices 1-4 Q = 5000 cfs

2. 7 7.5 Sluices 5 -8 Q = 5000 cfs

4.2 9 . 6 Sluices 1-8 Q = 10,000 cfs

Intermediate wall 490 150 7 - 35 2 . 0 - Existing outer wall reinstalled with 0.9 Sluices 1-4 Tunnel extension Q = 5000 cfs in place 1. 0 Sluices 5-8

Q = 5000 cfs 490 75 7 - 6.8 large - Existing outer wall

3. 7 25 - Sluices 1-4 Q = 5000 cfs

4. 3 9.4 - Sluices 5-8 Q = 5000 cfs

490 230 7 - 75 3.5 - Sluices 1-4

82 2.6 - Q = 1\000 cfs-Sluices -8

Q = 10 000 cfs

19

SUMMARY TABLE OF RESULT S (Continued)

Wa ter Tot al Model Sediment Dis c harge feed rate of in % of

Pond in 1000 cfs Sediment Total Model See Scheme level River Haveli T-S H/hr Haveli T-S Remarks Fig.

A Modifica- 490 75 7 12 4, 3 38 1 3 Existing outer wall tion III 13 79 11 Sluices 1- 4

Q = 5000 cfs 12 7 12 Sluices 5 -8

Q = 50 00 cfs 490 230 7 12 6 1 2 9 Sluices 1-4

Q = 10,000 cfs 57 3, 2 11 Sluices 5-8

Q = 10,000 cfs Extens ion of 490 230 7 12 55 2 . 4 7,4 Sluices 1-4 Tunnels 1 a nd Q = 10,000 cfs 2 only 2,5 11 , 3 Sluices 5-8

Q = 10,000 cfs 490 150 7 12 39 1, 9 9 , 4 Exis t ing outer wall

35 4,0 8 . 5 Sluices 1-4 Q = 5, 000 cfs

35 5,5 10, 1 Sluices 5-8 Q = 50 00 cfs

490 75 7 12 17 70 8.8 E xisting outer wall 16 50 9 .5 Sluices 1-4

Q = 50 00 cfs 21 9 , 8 6. 7 Sluices 5-8

Q = 50 00 cfs Existing 490 75 7 - 13 1. 4 - Existing conditions Conditions 11 7,8 Sluices 1-4

Q = 50 00 cfs 12 1. 6 Sluices 5-8

Q = 5000 cfs 490 150 7 - 24 12, 8 - Existing conditions

28 1. 9 Sluices 1-4 Q = 5000 cfs

24 5, 0 Sluices 5-8 Q = 50 00 cfs

490 225 7 - 58 2,2 Sluices 1-4 Q = 5000 cfs

52 7 , 0 Sluices 5-8 Q = 50 00 cfs

20

I-z w ~

EXTENSION OF TUNNELS MOO . EXISt EXPLANATION 0 w en _J

8 _J <1'. wz 7 - i" 0<1'. 6 OU ~ 5 _J a:: 4 <1'. w 1-I-o z 3 I- w

- :J w -> <1'.

_I

--

LL I 2 0 S2

--

I- I Z:i:= 0 w u A a:: w a.

IS.2 ONLY lII

en en en en I I I I I I I I I- I- I- I-

I B C D A

ISLANB ,MOVED GUI E ALL INSTALLED

CO\ID.

en I I I-

I

--->----

8 7 6 5 4 3 2 I

0

A. Existing outer divide wall B. Outer wall 155° long C. Existing outer wall sluices

I - 4 with Q = 5000 c.f.s.

D. Existing outer wall sluices 5-8 with Q = 5000 cf.s. sluices 1- 4 c losed

Fig. 24 COMPARISON OF EXTENSION OF TUNNELS I AND 2 ONLY WITH SCHEME A MODIFICATION ][

The continuous flow of wat er through the sluice gat es does not , in general , reduce sediment entry into the canals . The greater amounts of water flowing toward the head regulators cau ses gr eat er velocities which in turn creat es more suspension of bed material in the flow. Thus while s luicing pre -vents deposition of sediment in the immediate

vicinity of the head regul ators it does cause great er amount s of sediment to be transported into the canals . It is better to construct a s tructure which will dir ect the bed load toward the barrage away from the head regul at ors, and provide sluicing devices t o remove s i zaule deposits of sediment which will undoubt edly occur in areas of reduced velocities .

RECOMMENDATION

It is r ecommended that ~cheme A, Modifica-tion III shown in plan in Fig . 25 be const ructed as th e most suitab le among those tested to prevent excessive sediment entry into the canals . The recommended s tructure by its geometry creates a curve in the approach flow favorable to the canal head regulators. T he deflection of flow by the silt exclusion s tructure creat es a secondary current and movement of a substantial amount of bed load away from the head regulator t oward the barrage to the right of the outer divide wall.

The intermediate divide wall should be re -moved and it is recommended that the outer divide wall remain unc hanged . A shorter outer divide wall will decrease the sediment entry into the T - S

link but will increase, in gen eral, sediment ent ry into Haveli over its present condition. It is recommended further t hat the i s land pres ently obstructing the approach to the cana ls be r emoved in part or a total and the l eft guide wall of the kind recommended from results of the Pakistan model s tudy be provided. One s light modification to this guide wall is suggested at the left s ide of the T - S head r egulator. A wing wall rounded in the form shown in Fig. 16 should be con-st ruct ed to provide smoother flow entrance in the first and second bays of the head regulator .

The cente rline of the head regulators of the two canals s hould be construct ed in the same line with a distance of 330 ft between canal centerlines .

ALTERNATIVE RECOMMENDATION

Should the recommended st ructur e be consider-ed excessive in cost or otherwise difficult to construct and is not acceptable on these accounts , the alterna-tiv e scheme recommended for construction is exten -sion of tunnels one and two only as shown and described in Fig . 22. T he intermediat e guide wall should be removed and the outer guide wall should r emain intact T he P akis t an recommended left guide wall in the

2 1

approach region i s satisfactory provided that a wing wall of Fig . 25 is constructed adjacent to the l eft side of the T-S head regulator .

This alternative structure will cause great er sediment entry into the Haveli Canal than the recommended scheme .

N N

I I 11 .~~ • I 194° I

:xJ II -0 0 0

/ EL . 468

CHE NAB RIVER

FIG. 25

' ~<t TRIMMU-SIDHNAI I HEAD REGULATOR

330

ct HAVELI CANAL

ct TRIMMU-SIDHNAI / BRIDGE

<t. HAVELI REGULATOR --~-.......... ---~ SRI OGE

__f EL. 481 ------------------ -- ---- - - - - - - -- - - - - - - - - - - --~~:t~lt:::=== TOP OF SILT EXCLUDER

;;EL 468,_,_/ / EL. 468 z

EL. 468

EL.468~

RECOMMENDED SILT EXCLUDER

ACKNOWLEDGEMENTS

Acknowledgement is due Dr. D. B . Simons for his advice and participation in the model study and in the preparation of this report . The shop staff

of the Hydraulics Laboratory at Colorado State University constructed the model and assisted in the operation throughout the s tudy.

BIBLIOGRAPHY

Allen, J. Scale Models in Hydraulic Engineering, Longmans, Green and Co., London, 1947.

Einstein, Hans A ., and Hing-Chien, Similarity of Distorted River Models with Movable Beds, Transactions ASCE, Vol. 121, 1956.

Nichols , K . O. , Geometric Distortion in Hydraulic Models, Paper No. 2044 , Transactions ASCE , Vol. 104, 1944

23

Blench, T., Scale . Relations Among Sand - Bed Rivers Including Models , Proceed ings Se . No . 667 , ASCE , April 1955 .

Hydraulic Models, ASCE Manual of Engineering Practice No . 25, July 1942.